def get_board_info(board_type='synthetic', connection='usb', usb_port=None):
""" Gets the BarinFlow ID for the respective OpenBCI board in use
Parameters:
board_type (str): Type of base OpenBCI board (options: synthetic, cyton, daisy)
connection (str): Connection method, either via USB dongle or the Wifi board
Returns:
board_id (int): Id value for the board in the BrainFlow API
"""
params = BrainFlowInputParams()
if usb_port is None:
usb_port = STD_PORT
if board_type == 'synthetic':
board_id = BoardIds.SYNTHETIC_BOARD.value
elif board_type == 'cyton':
if connection == 'usb':
board_id = BoardIds.CYTON_BOARD.value
params.serial_port = usb_port
elif connection == 'wifi':
board_id = BoardIds.CYTON_WIFI_BOARD.value
elif board_type == 'daisy':
if connection == 'usb':
board_id = BoardIds.CYTON_DAISY_BOARD.value
params.serial_port = usb_port
elif connection == 'wifi':
board_id = BoardIds.CYTON_DAISY_WIFI_BOARD.value
return board_id, params
def __init__(self, board_id=BoardIds.CYTON_DAISY_BOARD.value, ip_port=6677, serial_port="COM3"):
# Board params
self.board_id = board_id
self.params = BrainFlowInputParams()
self.params.ip_port = ip_port
self.params.serial_port = serial_port
self.board = BoardShim(board_id, self.params)
self.sfreq = self.board.get_sampling_rate(board_id)
self.marker_row = self.board.get_marker_channel(self.board_id)
self.eeg_names = self.board.get_eeg_names(board_id)
# Features params
# todo: get as arg
self.features_params = {'channels': ['C03', 'C04']}
def run(self): # 线程执行函数
params = BrainFlowInputParams()
board_id = self.board_set[0]
self.c_value = []
params.serial_port = self.board_set[1]
self.board = BoardShim(board_id, params)
# 在线数据
#一导联数据和量导联数据
self.board.prepare_session()
self.board.start_stream()
time.sleep(3)
n = 0
m = np.zeros(2)
while self.is_on:
time.sleep(1)
data = self.board.get_current_board_data(self.board_set[3])
data = data[1:9]
mean_data = np.tile(
data.mean(axis=1).reshape(8, 1), (1, self.board_set[3]))
# print(data.shape)
data = data - mean_data
if self.puanduan1:
self.org_signal.emit(data)
filter1_data = self.processing_step1.step_1_list[
self.pre_set[0][0]](data, self.pre_set[0])
filter2_data = self.processing_step2.step_2_list[
self.pre_set[1][0]](filter1_data, self.pre_set[1])
if self.puanduan2:
# print(self.pre_set)
self.pre_signal.emit(filter2_data)
feature_data = self.feature.feature_list[self.class_set[0]](
filter2_data, self.class_set)
# print(np.argmax(feature_data))
if self.puanduan3:
self.feature_signal.emit(feature_data)
# print(feature_data)
result = np.argmax(feature_data)
r = feature_data[result]
# 更改部分（将程序中2s读取一次改为1s读取一次，每次512采样点改为256采样点）
if result == 5:
r = r + 0.1
elif result == 6:
r = r + 0.05
if r > 0.5:
n = n + 1
p = result + 1
print("分类结果：", p, " n = ", n)
m = np.append(m, p)
m = m[1:]
if m[0] == m[1]:
m[1] = 0
else:
b = str(p)
self.command_signal.emit(b)
print("发送指令：", p)